Preterm birth is a leading cause of infant mortality and morbidity in the U.S. and across the globe. Infection has consistently been identified as an important risk factor for preterm birth. A two to three-fold increase in the risk of preterm delivery is associated with the occurrence of reproductive tract infections during pregnancy. The overall goal of our study is to understand the role of microbes in the vaginal and oral cavities in triggering preterm birth, and how this role may be modified by human genotype and known behavioral risk factors. We propose using already collected specimens from a prospective cohort study of 354 African American women in Baltimore;16% of the cohort pregnancies resulted in a preterm birth. We also will test a/ready collected vaginal specimens from 300 women who delivered prematurely and 300 who did not, who participated in the Vaginal Ultrasound Cerclage Study. Our study will address the following hypotheses: 1. Preterm birth is associated with the presence of selected microbes found in the vaginal cavity, oral cavity, or both. 2. The association of microbes with preterm birth is modified by host characteristics. While there is evidence from a variety of different studies suggesting that both hypotheses are true, the joint effects of vaginal and oral microbes on preterm birth have yet to be addressed. For this proposal we have the great advantage of exploring these hypotheses using already collected specimens and data as part of completed studies of preterm birth.